Research Scientist Intern, NMR Analysis Automation

Meta•Redmond, WA

42d•$7,313 - $12,134

About The Position

Meta Reality Labs is seeking a Research Scientist Intern to contribute to cutting-edge materials research in support of the next generation of hardware. Accelerating battery materials discovery requires evaluating large numbers of electrolyte formulations, but analytical characterization remains a manual bottleneck. As an intern, you will build an end-to-end automated NMR analysis pipeline that transforms raw spectrometer data into ML-ready descriptors, reducing expert analysis time from hours to seconds per spectrum while maintaining quantitative accuracy. Your work will directly enable closed-loop, AI-driven electrolyte discovery for next-generation batteries.

Requirements

Currently pursuing a PhD in analytical chemistry, physical chemistry, chemical engineering, or a related field
Experience with solution-state NMR spectroscopy – theory, data interpretation, and quantitative analysis
Proficiency in Python for scientific data processing, including experience with spectral analysis or signal processing
Experience building automated data processing pipelines or batch analysis workflow

Nice To Haves

Experience with 2D (COSY) NMR and pulse field gradient NMR experiments
Experience designing experiments and analyzing data to validate hypotheses (e.g., DOE, statistical analysis, reproducibility studies)
Familiarity with machine learning concepts and feature engineering for predictive models
Background in electrochemistry or lithium-ion battery materials
Familiarity with Bruker hardware and software, including raw data formats
Experience with NMR of battery electrolytes, ionic liquids, or concentrated salt solutions

Responsibilities

Design and execute data processing pipelines for automated analysis of NMR spectra of liquid non-aqueous battery electrolytes (1H, 7Li, 13C, 19F, 31P nuclei) with a clear, measurable speedup over manual analysis
Implement signal processing workflows including denoising, automated peak identification and assignment, and constrained deconvolution for overlapping spectral regions in multi-component electrolyte mixtures
Identify the optimum parameters of the pipeline that balance accuracy and throughput
Convert NMR spectra into molecular level compositional and structural descriptors (concentrations, chemical structures) to be used for machine learning models
Build spectral quality assessment modules that flag problematic spectra and unreliable fits